AperTO - Archivio Istituzionale Open Access dell'Università di Torino
Original Citation:
Changes in soil mineral N content and abundances of bacterial communities involved in N reactions under laboratory conditions as predictors of soil N availability to maize under field conditions
Published version:
DOI:10.1007/s00374-016-1095-7 Terms of use:
Open Access
(Article begins on next page)
Anyone can freely access the full text of works made available as "Open Access". Works made available under a Creative Commons license can be used according to the terms and conditions of said license. Use of all other works requires consent of the right holder (author or publisher) if not exempted from copyright protection by the applicable law. Availability:
This is the author's manuscript
This full text was downloaded from iris - AperTO: https://iris.unito.it/
This is the author's final version of the contribution published as:
Fiorentino, Nunzio; Ventorino, Valeria; Bertora, Chiara; Pepe, Olimpia;
Giancarlo, Moschetti; Grignani, Carlo; Fagnano, Massimo. Changes in soil
mineral N content and abundances of bacterial communities involved in N
reactions under laboratory conditions as predictors of soil N availability to
maize under field conditions. BIOLOGY AND FERTILITY OF SOILS. 52
(4) pp: 523-537.
DOI: 10.1007/s00374-016-1095-7
The publisher's version is available at:
http://link.springer.com/content/pdf/10.1007/s00374-016-1095-7
When citing, please refer to the published version.
Link to this full text:
Changes in soil mineral N content and abundances of bacterial communities involved in N reactions under
1
laboratory conditions as predictors of soil N availability to maize under field conditions
2
3
Nunzio Fiorentino1*, Valeria Ventorino2, Chiara Bertora3, Olimpia Pepe2, Moschetti Giancarlo4, Carlo Grignani3, Massimo
4
Fagnano1
5
6
1Department of Agriculture, Division of Plant Biology and Crop Science, University of Naples Federico II, via Università,
7
100, 8055 Portici, Italy
8
2Department of Agriculture, Division of Microbiology, University of Naples Federico II, via Università, 100, 8055 Portici,
9
Italy,
10
3Department of Agriculture and Forest sciences, University of Turin, Largo Paolo Braccini, 2, 10095 Grugliasco, Italy
11
4Department of Agricultural and Forest Science, University of Palermo, Viale delle Scienze, Bldg 4, 90128 Palermo, Italy
12
13
*Corresponding author.
14
Department of Agriculture, Division of Plant Biology and Crop Science, University of Naples Federico II, via Università,
15
100, 8055 Portici, Italy16
Phone: +39 081253912917
Fax: +39 081253915518
email: nunzio.fiorentino@unina.it19
Abstract
20
Proper management of soil fertility requires specific tools for predicting N availability for crops as a consequence of different
21
fertilization strategies. More information is required, especially for organic fertilizers, depending on their mineralization rate,
22
composition and processing (i.e., fresh or composted manure), as well the influence of soil characteristics on their effect.
23
Laboratory soil incubations were used as a proxy for understanding plant-soil N dynamics under field conditions. Chemical
24
and microbiological measurements as contents of mineral N, potentially mineralizable N and the abundance of key genes
25
regulating the overall N-cycle were used as predictors of mineral N availability to maize in two contrasting pedoclimatic
26
conditions. Our results showed that there was a good correlation between chemical and microbiological measurements from
27
laboratory soil incubation experiments and soil-plant N dynamics of maize cropping systems. Mineralization patterns from
28
soil incubation proved to be useful for optimizing fertilization management of maize under field conditions as long as
29
incubation time is normalized over maize growth cycle, according to the simplified model of Growth Degree Days. Average
30
cumulative soil mineral N values calculated over a short incubation period (42 days) showed a significant correlation
31
(R2=0.72, p<0.05) with maize N uptake. The shape and kinetic parameters of net N mineralization from medium term (112
32
days) soil incubation provided consistent information on the interaction between fertilizers and native fertility. The abundance
33
of N fixation, nitrification, and denitrification genes (nifH, amoA, nirK and nirS) was sensitive to soil characteristics and N
34
fertilization. This work provides a suitable starting point for developing a crop-based approach for using incubation data to
35
optimize maize fertilization. However, more studies with different maize cultivars and pedoclimatic conditions are needed to
36
generalize this approach.
37
38
Keywords: N mineralization; curve fitting; Growth Degree Days; nifH; amoA; nirK
Introduction
40
The decline in soil organic matter (SOM) content is a major threat to soil fertility in European semi-arid regions due to their
41
specific climatic conditions (Diacono and Montemurro 2010). Therefore, proper fertility management must be utilized to
42
build up SOM so that the soil can cope with the high mineralization rate of these environments. Among these strategies,
43
fertilization with composted manures provides a good option in areas without manure from livestock farms with positive
44
effects on soil chemical and physical properties as the increase in OM sequestration (Gabrielle et al. 2005; Fagnano et al.
45
2011; Alluvione et al. 2013), aggregate stability (Diacono and Montemurro 2010; Chen et al. 2014; Meng et al. 2014) and
46
hydraulic conductivity (Chen et al., 2014), which are primarily due to the hydrophobic protection of stabilized SOM (Spaccini
47
et al. 2002; Piccolo et al. 2004; Lynch et al. 2006). Moreover, compost fertilization can increase soil biological activity and
48
microbiological diversity (Ventorino et al. 2013), as well as suppress soil-borne diseases (Ventorino et al. 2016). Although
49
compost maturity level (Annabi et al. 2007) and its chemical composition (Miller et al. 2010) are recognized as key factors in
50
improving soil physico-chemical characteristics, climatic variables and soil properties may explain the high variability in
51
mineral N availability of compost amended soils (Nedvěd et al. 2008). The highest mineralization rates of compost occur in
52
high pH soils (Huang and Cheng 2009). Soil texture is another key factor controlling N mineralization of composted manures
53
(Saenger et al. 2014), as for example NH4+ is retained by ammonium fixing clays (Nyamangara et al. 1999) and this decreases
54
nitrification.
55
Knowledge of the fate of N from organic fertilizers is required for optimizing N rates to obtain valuable crop
56
performances and minimize the impacts of N surplus on the environment (Zebarth et al. 2009). Soil incubation has been used
57
to study N patterns in manure-amended soils (Cordovil et al. 2005; Flavel and Murphy 2006), fitting mineral N data to
58
functional models to estimate pools of mineralizable organic N (De Neve and Hofman 2002, Griffin et al. 2005). These studies
59
have obtained useful insights into mineral N availability from fertilizers. Riberio et al. (2010) reported a positive correlation
60
between estimated N mineralization and lettuce N uptake. Abbasi et al. (2012) found that maize biomass and N uptake were
61
correlated. Nevertheless, all of these results are from pot experiments, and caution is required when extrapolating the results
62
of these pot experiments to field conditions.
63
The N cycle includes key processes such as N-fixation, nitrification and denitrification (Ventorino et al. 2010; Weng
64
et al. 2013), and microorganisms involved in these processes can be detected by molecular techniques through the
65
quantification of genesencoding enzymes such as nitrogenase reductase (nifH), ammonia monooxygenase (amoA) and nitrite
66
reductase (nirS and nirK). Among available molecular methods, quantitative real-time polymerase chain reaction (q-PCR) is
the most sensitive and most suitable approach for determining abundance of functional genes from soil-derived DNA and
68
RNA (Sharma et al. 2007).
69
In this work, we used data from a three-year field study (Alluvione et al. 2013) and a medium term (112 days) soil incubation
70
period to find the relationships between field and laboratory experiments in studying N availability in fertilized soils. Chemical
71
and bacterial properties, such as contents of mineral N, potentially mineralizable N and the abundance of the aforementioned
72
key genes in soil, were determined for predicting mineral N availability for crops in the field.
73
Our work aimed to answer the following questions: (i) are results from field and laboratory experiments related?; (ii) is it
74
possible to identify a simple index from laboratory incubations that can predict crop performance and optimize fertilization
75
strategy?; (iii) can the evaluation of the abundance of key genes of N-cycle be used for predicting N available for plant
76
nutrition?
Material and methods
78
79
Experimental set-up80
81
Field trials were carried out over three consecutive years (2006-2008) under two contrasting pedoclimatic conditions to assess
82
the effects of different types of fertility management on maize crop yield and soil C and N dynamics. A detailed description
83
of the experimental set-up, sampling techniques, chemical analyses and data elaboration is reported by Alluvione et al. (2013).
84
Briefly, sites were located in the Po River Valley near Torino (TO site - 44°53' N, 7°41'E, Piedmont Region, Northern Italy)
85
and in the Sele River Plain near Napoli (NA site - 40°37' N, 14°55' E, Campania Region, Southern Italy). In this work, we
86
report yield and soil mineral N content of the two sites concerning the following treatments: compost (COM) and urea (UREA)
87
fertilization added to soil at the same N rate (130 kg N ha-1) compared to an unfertilized control (0N). Compost was made
88
with a park, garden, and separately collected urban waste mixture and, according to Alluvione et al (2013), had the following
89
composition: 610 g kg-1 (f.m.) dry matter; 21 g kg-1 (d.m.) ash; 256 g kg-1 (d.m.) organic C; 391 soluble C / total C ratio; 12.2
90
g kg-1 (d.m.) total N. Compost was not phytotoxic at the 150 Mg ha−1 rate used for the plant growth assay proposed by
91
Chukwujindu et al. (2006). Treatments were conducted in quadruplicate following a completely randomized design at the TO
92
site, while a randomized complete block design was used at the NA site.
93
The soil incubation experiment aimed to investigate mineral N and bacterial properties of the two soils and to study
94
the relationship of these two measurements under controlled laboratory conditions.
95
Soils were collected (0-30 cm layer) at both field sites at the end of the 2nd year of the field experiment (autumn
96
2007) in an area near the experimental plots so as to have soils with the same properties as those used for the field experiment
97
but not affected by treatments. Soils were air-dried and sieved at 5 mm; the main soil characteristics are shown in Table 1.
98
Sieving size (<5 mm) was selected to minimize soil disturbance and preserve micro-aggregates. The following three factors
99
were arranged over a completely randomized design with three replicates: (i) 2 soils from Napoli (NA) and Torino (TO)
100
experimental fields; (ii) 3 types of fertilization: compost (COM), urea (UREA), and an unfertilized control (0N); and (iii) 8
101
sampling times. Totally, 144 experimental units (2 soils x 3 fertilizations x 8 samplings x 3 replicates) were set up, each
102
consisting of 400 g of dry soil in a 543 cm3 pot. Urea was added to soil as a liquid solution, while compost was grinded with
103
a hammer-mill before the application and individual soil-fertilizer mixtures were prepared for each pot. The quantities of urea
104
(0.08 g kg-1 soil) and compost (4.6 g kg-1 soil) to be added to each experimental unit were calculated taking as reference the
105
N field rate (130 kg N ha-1) and the 0-30 cm layer with an average bulk density of 1.2 Mg m-3 (measured after sieving and
soil consolidation with watering). A pre-incubation period of 5 days at 28°C and at field capacity was conducted to stabilize
107
soil microbial activity. Afterwards, a 112-day incubation was conducted in the dark at 28°C, controlling soil moisture to
108
simulate the fluctuations of ordinary maize irrigation. According to FAO 56 paper (Allen et al. 1998), irrigation events were
109
simulated at a threshold value of 45% of Water Holding Capacity in order to raise soil moisture to its field capacity.
110
Soil hydraulic parameters were determined in the laboratory by Richard’s chambers (Richards 1944). Air relative
111
humidity was kept between 60 and 70% to simulate the conditions of our field sites during the maize growth cycle.
112
113
Plant sampling and analyses
114
115
In the field experiment, total biomass was determined by hand-harvesting maize plants at the dent stage from an area of 15
116
m2 per plot. Dry matter was determined after oven-drying samples at 70°C until a constant weight was reached. The N content
117
of plant tissues was measured using a CHN elemental analyser (NA 1500 N analyser from Carlo Erba Instruments, Thermo
118
Fisher Scientific, Waltham, MA). Maize N uptake was calculated by multiplying N concentration in the entire plant by total
119
dry matter yield.
120
121
Soil sampling and analyses
122
123
In the field experiment, soil samples were collected from three soil layers (0-15, 15-30, 30-60 cm) before sowing (April), at
124
flowering (June), and after harvest (October) of each cropping cycle. In the laboratory incubation experiment, three destructive
125
soil samples per each soil x fertilization combination were randomly collected at 0, 7, 14, 28, 42, 63, 84 and 112 days for
126
chemical analysis. Concentrations of NH4+-N and NO3--N were measured according to the Hach® method and the extracts
127
were analysed by spectrophotometry (Hach DR 2000, Hach Company, Loveland, CO). This method was used in both field
128
and incubation experiments with the exception of soil samples from the TO field experiment. In this case, mineral N was
129
extracted by 1 M KCl and analysed with a continuous flow analyser (Evolution II, Alliance Analytical Inc., Menlo Park, CA).
130
131
Quantification of bacterial functional genes
132
133
Soil samples were collected from pots after 42 and 112 days of incubation as reported above. Total DNA was extracted by
Bacterial functional genes were quantified with qPCR. Amplification of nifH, amoA, nirK and nirS genes were
136
conducted in a Chrom4 System Thermocycler (Bio-Rad, Milano, Italy) using the primers listed in Table 2. The PCR mixture
137
(25 µL total volume) included 50 ng of target DNA, 1× iQ SYBR Green supermix, (Bio-Rad) and 0.2 µM of each primer. All
138
qPCR reactions included an initial denaturing step at 95°C for 3 min followed by 40 cycles of amplification as specified in
139
Table 2. Standard curves were generated to calculate the abundance of evaluated genes. Each target gene was amplified with
140
primers from different type strains: nifH (nifH-F, AAAGGYGGWATCGGYAARTCCACCAC-3’; nifH-R,
5’-141
TTGTTSGCSGCRTACATSGCCATCAT-3’) and nirK (nirK-876, 5’-ATYGGCGGVAYGGCGA-3’; nirK-5R, 5’-
142
GCCTCGATCAGRTTRTGG-3’) from Sinorhizobium meliloti 1021; amoA (amoA-1F, 5’-GGGGTTTCTACTGGTGGT-3’;
143
amoA-2R, 5’-CCCCTCKGSAAAGCCTTCTTC-3’) from Nitrosomonas europea; and nirS (nirS-cd3aF, 5’-
144
AACGYSAAGGARACSGG-3’; nirS-R3cd, 5’-GASTTCGGRTGSGTCTTSAYGAA-3’) from Pseudomonas stutzeri
145
DSM5190. Amplicons were purified from agarose gels using a QIAquick gel extraction kit (Qiagen S.p.A, Milano, Italy) and
146
cloned using TOPO-TA cloning kit dual promoter (Invitrogen, Milano, Italy) in Escherichia coli competent cells according
147
to the manufacturer’s protocol. Plasmid DNA containing a single copy of the target gene was purified with QIAprep Spin
148
Miniprep Kit (QiagenS.p.A) and quantified using the QuantiFluorTM (Promega, Milano, Italy). The copy number per microliter
149
of each cloned gene was calculated using the following formula (Chen et al. 2015): (A×B)/(C×D) where A is the concentration
150
of the DNA template (ng µl-1), B is the Avogadro number (6.023×1023 copies mol-1), C is the average molecular weight of a
151
DNA base pair (6.6×1011 ng mol-1), and D is the DNA size (bp). Ten-fold serial dilutions, ranging from 101 to 108 gene copies
152
µl-1, of each plasmid DNA sample were used to generate a standard curve for each target gene (nifH, amoA, nirK, and nirS).
153
154
Data elaboration and statistical analyses
155
156
We conducted Kolmogorov–Smirnov and Levene tests to verify the assumptions of normality and homoscedasticity of the
157
data. For field data, we used a general linear model of analysis of variance (ANOVA). Maize N uptake, soil NH4+-N, and
158
NO3--N contents from field experiments were analysed using fertilization (0N, COM and UREA) as the main factor, while
159
year was treated as a repeated measure. We verified the variance-covariance structure of our dataset by performing Mauchly’s
160
sphericity test. Field data were analysed separately per site to preserve homogeneity of variance. Soil mineral N content was
161
transformed through natural logarithm (of values added to 100 to eliminate 0 values) to preserve normality of the data before
162
conducting the ANOVA.
For laboratory incubation, we analyzed NH4+-N, NO3--N and total mineral N content data using a one-way ANOVA,
164
which included three factors (soil x fertilization x sampling time). Next, a one-way ANOVA with one factor was used, which
165
considered each soil x fertilization x sampling combination as a single treatment to compare soil mineral N values recorded
166
at each sampling date. As for field data, soil mineral N content was natural logarithm transformed (of values added to 100 to
167
eliminate 0 values) to preserve normality of the data before conducting the ANOVA.
168
For microbiological data, we used a one-way ANOVA with three factors (soil x fertilization x sampling time).
169
All statistical analyses were performed using the SPSS 13.0 statistical software package (SPSS Inc., Cary, NC, USA);
170
statistical comparisons were made at the α = 0.05 probability level and mean values were separated using the Sidak post-hoc
171
test.
172
According to Abbasi and Khizar (2012) and Ribeiro et al. (2010), cumulative soil mineral N values from soil
173
incubations can be used to predict plant uptake. We tested this hypothesis by plotting cumulative soil mineral N against
174
average maize uptake for three years of field experimentation. To plot these data, we converted soil mineral N values in kg N
175
ha-1 by using a reference layer of 0-40 cm and a bulk density of 1.2 Mg m-3.
176
Furthermore, an iterative curve fitting procedure (SigmaPlot 12.1, Systat Software, Inc., San Jose, CA) was
177
performed on cumulative net mineral N (n-SMN), which is the difference between each soil mineral N value and the value
178
recorded at day 0 for each treatment, with the aim of estimating the parameters of the following mineralization functions:
179
180
1) Single pool negative exponential function (Eq.1):
181
182
− () = × [1 − ( × )] (Equation 1)
183
where n-SMN (t) is the amount of N mineralized at time t, N is potentially mineralizable N and k is the decomposition
184
coefficient.
185
186
2) Two-pool negative exponential function (Eq.2):
187
188
− () = × [1 − ( × )] + × [1 − ( !" × )] (Equation 2)
where n-SMN (t) production is split into two N pools differing in their decomposability. The Nfast pool represents N in easily
190
decomposable organic materials (i.e., fresh residues), while Nslow pool includes more recalcitrant sources with a slow rate of
191
decomposition. The k fast and k slow are the corresponding decomposition coefficients of the two pools.
192
193
3) Sigmoid, 3-parameter function (Eq.3) :
194
195
− () = # $%&'() ()*) + (Equation 3)196
where Ns represents potentially mineralizable N, Ks is the inflection point, t0 is the time neededto raise SMN values to Ns/2,
197
and Ks is the mineralization coefficient.
198
In the Results section, we reported the best performing functions according to the regression coefficient (R2) values
199
and to the standard error of estimate (SEE). No constraints were applied to parameters estimated by the fitting procedure, and
200
the best performing functions were identified among those with realistic parameters for potentially mineralizable N (not higher
201
than soil total N) and the mineralization coefficient (between 10-3 and 1 day-1).
Results
203
204
Field experiment205
206
Nitrogen uptake of maize in the two sites is reported in Table 3. Unfertilized plots had higher N availability in TO soil, with
207
an average uptake that was two times greater than in NA soil (average value of 105.8 kg N ha-1 vs 181.5 kg ha-1 for NA and
208
TO site, respectively). Nitrogen uptake declined at both sites over the years, but in NA soil, this decrease was more marked.
209
In fact, taking values from the first year as a reference at NA, there was a dramatic decline from the second year (-40%), while
210
a comparable decrease (-43%) was observed in TO soil only in the third year as a consequence of hailstorm damage. Nitrogen
211
uptake at the TO site was significantly affected by COM and UREA fertilization in 2007 and 2008, with an average increase
212
of +34% for fertilized plots with respect to 0N. In contrast, compost fertilization dramatically reduced N uptake at the NA
213
site (-59% on the average), while UREA produced a 74% increase on average. Nitrogen uptake values recorded with UREA
214
were comparable between the two sites, even though values significantly declined from the first to the third year at the NA
215
site (-54% decrease), whereas TO soil was able to sustain N nutrition of maize with an N uptake always above 200 kg N ha
-216
1. The addition of compost did not affect N availability in TO soil with the exception of year 2008 (-29% compared to UREA
217
fertilized soil) while it strongly reduced N uptake in NA soil (from -83% to -60% compared to the UREA fertilized soil). Soil
218
mineral N content recorded in the 0-60 cm layer was almost three times higher in NA than in TO soil both at the pre-seeding
219
(Fig. 1a) and the harvest (Fig. 1b) stage. At the pre-seeding stage, differences among the treatments only occurred in NA soil,
220
with lower values recorded for 0N and COM (on average -13% with respect to UREA), while TO soil showed the same pattern
221
at harvest. In this stage no differences were recorded for mineral N in NA soil even though the UREA effect was significant
222
for NH4+-N values (p=0.027) with COM values -49% lower than those of UREA. Nitrate-N content followed essentially the
223
same pattern of soil mineral N, while NH4+-N content was almost constant in both soils and ten times higher in NA (59 kg N
224
ha-1, average value) than in TO (6 kg N ha-1) soil. The NH
4+-N:NO3--N ratio was significantly higher in NA (approximately
225
0.45) than in TO (0.18) soil (data not shown).
226
227
Incubation experiment
228
Soil mineral N patterns and curve fitting
229
A significant soil x fertilization interaction was detected (p<0.01) by a one-way three factor ANOVA (soil x fertilization x
230
sampling time), as shown in Figure 2a. As already found for maize N uptake (see Table 3), the 0N treatment had the largest
231
difference between the two soils with soil mineral N values significantly higher in TO soil (+56%). Compost fertilization did
232
not affect soil mineral N content in NA soil compared to soil of the 0N treatment, while a significant reduction (-18%
233
compared to 0N) was observed for TO soil, even though values were still higher than those of the same treatment in NA soil.
234
The UREA treatment increased mineral N content of both soils, but its effect was stronger for NA soil (+82%) than for TO
235
(+20%) soil respectively. The NH4+-N: NO3--N ratio of the 0N treatment (Fig. 2b) was comparable to that found in the field
236
experiment (0.35 and 0.18, in NA and TO soil, respectively). The UREA treatment significantly reduced this ratio in both
237
soils (0.16 on average). Compost reduced this ratio in both soils, but the differences were only significant in the TO soil. Net
238
N mineralization patterns of NA and TO soils are reported in Figures 3 and 4, showing that normalized soil mineral N
(n-239
SMN) patterns for each soil x fertilizer combination together with the best fitting function (see Materials and methods section).
240
For NA control soil (Fig. 3a), n-SMN values did not show variation until day 42, at which point, n-SMN values significantly
241
increased and reached a plateau at 84 days (40 mg kg-1). Compost fertilization (Fig. 3b) reduced the length of the initial 0 mg
242
kg-1 constant phase by 21 days, then there was a double step increase, the first at 28 days (from 3 to 13 mg kg-1) and the second
243
between 42-112 days (from 17 to 50 mg kg-1). Both 0N and COM treatments are well described by a sigmoid function (R2 =
244
0.98), although compost fertilization increased the potentially mineralizable N from 39 to 51 mg kg-1, reducing the initial
245
steady state. In the UREA treatment, n-SMN rapidly increased (Fig. 3c) during the first 2 weeks (until 30 mg kg-1), then a
246
new increase was recorded between 42-63 days, reaching a maximum value of 62 mg kg-1. This treatment modified the shape
247
of the curve to a double exponential rise (R2 of 0.98 vs 0.88-0.87 of the other models), with a recalcitrant N pool of 82 mg kg
-248
1, a mineralization rate of 0.01 mg kg-1 N day-1 and an easily mineralizable N pool of 12 mg kg-1 (k=0.22 day-1), likely
249
representing urea N. In the TO control soil (Fig. 4a), n-SMN values increased by approximately 20 mg kg-1 during the first
250
week and then increased again from day 28 to the end of the experiment, reaching a maximum value of 57 mg kg-1. The
251
addition of compost (Fig. 4b) changed this pattern with a double step increase between 7 -14 days (from 2 to 22 mg kg-1), and
252
between 42-63 days (from 20 to 38 mg kg-1), getting a final cumulative value of 43 mg kg-1. The UREA fertilization (Fig. 4c)
253
rapidly increased n-SMN values, producing 62 mg kg-1 of mineral N in the first 63 days.
254
Mineral N of TO soil was perfectly fitted with a single pool exponential function. Fertilization was able to modify
255
the two parameters of the function coherently with the quality of the N input. Compost fertilization did not produce any
256
significant variation in the kinetic parameter (k=0.019 and 0.021 day-1 for 0N and COM, respectively) while reduced the
257
mineralizable N pool (from 65 to 47 mg kg-1).
The UREA treatment did not change the mineralizable pool (almost 66 mg kg-1) and slightly increased the mineralization rate
259
(from 0.019 to 0.022 day-1).
260
261
Mineral N incubation data as a proxy for field N uptake
262
We tested the correlation between mineral N content of lab-incubated soils and the average N uptake (3 years) of maize from
263
our field experiment. As shown in Figure 5a, cumulative soil mineral N values at day 112 explained only 30% of the variability
264
of the data. Because of these findings, we propose a new crop-based approach for using laboratory incubation data, starting
265
from the hypothesis that incubation time must be normalized over the cycle of the target crop to be studied. Therefore, we
266
converted the incubation time in maize thermal time, that is, the sum of heat units measured in Growth Degree Days (GDD)
267
(Pampana et al. 2009). The most critical phase for maize N requirements is generally that between emergence and silking.
268
For the maturity class used in our experiment (FAO 500), this phase corresponds to 785.5 GDD and is equal to almost 42
269
days of incubation at a daily constant temperature of 28°C. Therefore, we plotted the cumulative soil mineral N value at 42
270
days (Fig. 5b) against the average N uptake (3 years) of our field experiment and obtained a regression coefficient of
271
approximately 0.582 and a significance close to 0.06 (d.f=4). Moreover, the performance of the regression increased
272
significantly when average soil mineral N of the first 42 days was used and it accounted for 72% of maize N uptake (p<0.05)
273
variability recorded in the field (Fig. 5c).
274
275
Functional gene analysis
276
A significant difference in the abundance of N-fixers was observed in the studied soils, with values of nifH gene abundance
277
significantly higher in NA than in TO soil (p < 0.05) (Fig. 6 a). Moreover, a tendency to decrease after 112 days was recoded
278
for both soils, even though significant variation was only detected for 0N and COM treatments in the NA soil (p<0.05).
279
Compost only significantly increased the abundance of N-fixers in the NA soil at 42 days (from 7.20 x 104 to 2.28 x 105 copies
280
g-1 soil, for COM and 0N, respectively). In contrast, inorganic fertilization significantly reduced nifH gene abundance at both
281
sampling times in NA and TO soils, having the lowest values of dinitrogenase reductase copy numbers, ranging from 1.10 x
282
103 to 5.31 x 103 (Fig. 6a).
283
The effect of soil and fertilization on ammonia-oxidizers showed an inverse pattern with respect to N-fixers (Fig.
284
6b). At 42 days, a lower (but not significant) abundance of the amoA gene was recorded in NA soil treated with 0N and COM
285
and a significant increase in gene copy number (1 to 2 orders of magnitude) in both UREA-treated soils (with the exception
of NA soil at day 42). This abundance was significantly higher in TO soil than in NA soil at both sampling dates with a
287
constant difference of half an order of magnitude.
288
The abundance of the nirK gene was influenced by soil type and fertilization (Fig. 6c). At 28 days, NA soil showed
289
significantly higher values than TO soil in 0N (2.38 x 105 vs 1.32 x 104) and COM (8.79 x 104 vs 1.32 x 104) treatments.
290
Fertilization reduced the abundance of the nirK gene in NA (approximately –37% lower than 0N). By contrast, in TO soil,
291
organic fertilization did not affect the abundance of denitrifiers (1.32 x 104 copies g-1 soil, on the average), while urea
292
fertilization significantly increased nirK gene abundance to a value of 9.98 x 104 copies g-1 soil (Fig. 6c). After 112 days, an
293
increase in nirK gene abundance was recorded especially in TO soil treated with 0N or COM, reaching values approximately
294
100-fold higher than those recorded at 42 days. In contrast, the abundance of the nirS gene was not affected by fertilization,
295
soil type and incubation period because no significant differences among pots were recorded (Fig. 6d). The denitrifier
296
community of both soils was dominated by nirK, whose abundance was at least 100-fold higher than that of nirS
297
(approximately 2 x 102 copies g-1 soil) (Fig. 6 d).
298
299
Discussion
300
301
Matching Field results with incubation data
302
303
The two soils differed in native fertility, as shown by crop performance in unfertilized plots (Alluvione et al. 2013). Crop N
304
uptake was significantly higher at the TO site, where it was slightly affected by fertilization, with an average value +85%
305
higher than that recorded in the NA site. In contrast, crop N uptake was sensitive to fertilization at the NA site.
306
Low pre-seeding soil mineral N values were recorded at the TO site and corresponded to high N uptake and vice versa at the
307
NA site; however, no correlation was detected between these two variables. These findings contradict those of
Sylvester-308
Bradley et al. (2001), who showed that soil mineral N content predicted crop performance and accounted for 78% of the
309
variation in wheat N uptake. Probably this discrepancy depends on the different environmental conditions between the
pre-310
seeding and the cropping season. Soil mineral N content of TO site during the pre-seeding season was low due to winter
311
leaching and soil coarse texture (Alluvione et al. 2013). On the contrary, soil mineral N values of the NA soil during the same
312
period were higher probably due to the higher clay content that limits fall-winter N leaching (Gaines and Gaines 1994).
313
The sigmoid soil mineral N pattern recorded with the incubation of unfertilized NA soil suggests that net N
314
mineralization requires 42 days to increase even under optimal conditions. Simard and N’Dayegamiye (1993) recorded the
same pattern in the first 40 weeks of soil incubations most likely because the initial N mineralization potential was more
316
affected by clay content (inverse proportionality) than by total C and N content (Zebarth et al. 2009). At the TO site, soil
317
mineral N values recorded at harvest in 0N plots were lower compared to pre-seeding values, suggesting that high crop N
318
uptake reduced N availability and that TO soil was able to reintegrate this N pool over the years. Moreover, despite the low
319
soil mineral N values recorded during the pre-seeding season, N uptake was always higher in TO soil than in NA soil,
320
suggesting that N mineralization was able to sustain crop requirements (even without fertilization).
321
In addition to increasing maize N uptake, UREA treatment significantly increased pre-seeding soil mineral N in the
322
NA soil, suggesting that easily available N inputs are mandatory for balancing the low N mineralization rate of this soil. This
323
finding is confirmed by incubation data where the UREA doubled the average soil mineral N of 0N treatment (from 30 to 60
324
mg kg-1) and modified the mineralization curve shape from sigmoidal to exponential plot, increasing the potentially
325
mineralizable N and limiting the initial delay phase. The UREA treatment did not increase N uptake in the TO soil but lead
326
to a residual soil mineral N value that was higher than that of the control. This suggests that even a low N fertilization rate,
327
compared to recommended N dose of 250 kg N ha-1 in these environments, is not needed to satisfy crop requirements. The
328
low sensitivity of TO soil to UREA fertilization is also demonstrated by the mineralization curve shape not differing from
329
that of unfertilized soil. In the first year, COM fertilization dramatically reduced maize N uptake in the NA soil, probably
330
because the first addition of compost rapidly reduced N availability, pushing soil fertility to its lowest limit. Average soil
331
mineral N at the pre-seeding stage, which did not differ from that of control plots, was significantly lower than that of
UREA-332
treated soils. This pattern has also been identified by other authors in low fertility soils (Eriksen et al. 1999; Hargreaves et al.
333
2008) and confirms data on N immobilization after fertilization with composted manures, even in soils with a C:N ratio of 9
334
(Nyamangara et al. 1999).
335
Thus, this finding indicates that there was an over-exploitation of soil N stock during the cropping season and a low
336
ability of NA soil to restore soil mineral N in the absence of easily available N inputs. Alluvione et al. (2013) hypothesized
337
that the low N availability of NA soil might be due to clay-mediated protection of soil organic N limiting N mineralization
338
rate. Nevertheless, this model does not completely match with incubation data: average soil mineral N values recorded from
339
the COM treatment did not show any reduction with respect to control soil, and mineralization pattern was positively affected
340
by the fertilization with a reduction in the initial steady phase. This difference was probably due to recurrent soil inversion by
341
tillage that increases soil compaction and anoxic sites, resulting in a severe limitation of SOM and compost mineralization
342
(Holland 2004). The controlled conditions of laboratory incubation, mainly less soil compaction, allowed for better soil
343
aerobic conditions necessary for compost degradation than field conditions, suggesting that in NA soil, organic fertilization
344
can have a positive effect if soil compaction is limited (i.e., adopting minimum tillage or sod seeding).
345
At the TO site, compost fertilization did not limit N uptake with the exception of the 2008. This difference was due
346
to the hailstorm damage that was only compensated by the treatment with UREA (Alluvione et al. 2013). Most likely, the
347
yield would have been the same among the treatments in the absence of this climatic event. Average soil mineral N values
348
recorded at maize harvest were lower in COM than in UREA soil, and a similar pattern was observed during incubation: the
349
average difference in soil mineral N was 20% and the decline of potentially mineralizable N from 65 to 47 mg kg-1. The
350
reduction in soil mineral N may be due to the reduced contribution of compost to maize nutrition (Amlinger et al. 2003) under
351
field conditions and N immobilization by soil microflora that occurred during incubation (Nyamangara et al. 1999). In any
352
case, maize performance was not limited due to the high SOM mineralization rate in the TO soil.
353
354
Soil laboratory incubations for predicting maize yield in the field
355
356
Soil mineral N production of NA soil (for both 0N and COM treatments) was close to zero during the first 42 days of
357
incubation, corresponding to a maize thermal time of 785 GDD that covers the period from emergence to silking. This virtually
358
means that during stem elongation, the mineral N content of soilwas not able to sustain crop growth also considering that
359
under field conditions, moisture and temperature could be not optimal for mineralization processes. The N mineralization
360
pattern during the incubation of NA soil explains why crop growth was lower than expected by considering soil mineral N of
361
the field experiment and suggests excluding the sole organic fertilization as a N source for maize under field conditions. In
362
contrast, TO soil showed a higher attitude to produce soil mineral N, with a fast increase during the first weeks that perfectly
363
fitted with a one pool negative exponential function. The unfertilized soil showed a cumulative soil mineral N of
364
approximately 56 mg kg-1 in 756 GDD, corresponding to approximately 268 kg N ha-1 in the 0-40 layer (B.D. of 1.2 Mg ha
-365
1). This indicates that TO soil is able to feed maize regardless of the type of fertilization management and explains why
366
unfertilized soil allowed a N uptake value close to 200 kg N ha-1 in the first two years of experiment. The significant correlation
367
(p = 0.06) between cumulative soil mineral N at 42 days and maize uptake suggested that short-term incubation can be used
368
to predict N uptake if the incubation time is synchronized with crop growth according to the concept of thermal units.
369
Averaging values during the 42 days periodincreased the performance of the correlation (p<0.05) most likely because mean
370
soil mineral N values can be considered as an integrated index of the different mineralization patterns occurring during the
371
incubation period. Cumulative soil mineral N at 112 days was a poor predictor of maize N uptake, meaning that medium and
long-term incubation is not a wise option when the aim is to estimate N availability in crops under field conditions using soil
373
mineral N data. Nevertheless, this approach is still the best choice for iterative fitting procedures aimed at finding the best
374
functional model to describe N mineralization in different soils.
375
376
Matching microbiological indices with soil chemical data from the field and laboratory experiment
377
378
The use of PCR to quantify the abundance of the functional and bacterial genes nifH, amoA, nirK and nirS encoding enzymes
379
involved in the N-fixation, nitrification and denitrification, respectively, is a sensitive method for evaluating the effect of
380
different soil types and fertilizers on the abundance of the relative bacterial communities. These measurements provide
381
indications on the effects of soil environmental factors, such as nutrients, oxygen status, pH, pollutants, agro-chemicals,
382
moisture, texture and temperature, on the abundance of these bacterial communities (Sharma et al. 2007). Our results confirm
383
that the abundance of these bacterial communities depends on soil proprieties (Delmont et al. 2014) and fertilization type.
384
N-fixing bacteria provide an important source of N to natural ecosystems (Hayden et al. 2010). Different physical
385
soil properties, such as texture, aggregate size and clay content, can influence N fixation in arable soils (Roper and Smith
386
1991; Poly et al. 2001). The higher abundance of diazotrophic bacteria in NA than TO soil highlights a positive correlation
387
between clay texture and the abundance of the nifH gene, confirming what reported by Pereira et al. (2013). Furthermore,
388
diazotrophic community composition has been shown to respond to changes in the type of the added N (Orr et al. 2011).
389
Organic fertilizers can stimulate N fixation and bacterial abundance (Keeling et al. 1998; Kondo and Yasuda 2003), while
390
available N can stimulate or inhibit both (Poly et al. 2001; Tan et al. 2003). Compost fertilization positively affected the
391
abundance of N fixers in NA soil at 42 days of incubation; however, the effect decreased over the incubation period, probably
392
as result of the available SOM added to the soil (Pepe et al. 2013).
393
In contrast, high N content in response to nitrate and/or ammonium added to soil may inhibit dinitrogenase reductase
394
and thus N fixation (Lindsay et al. 2010). The higher nifH abundance in NA soil than in TO soil, characterised by the lower
395
mineral N content, confirms this hypothesis. Additional confirmatory evidence comes from the fact that the lowest abundance
396
of diazotrophics was in the UREA treated soils. However, Rodriguez-Blanco et al. (2015) reported that the abundance and
397
composition of diazotrophic bacterial populations can be influenced by N-fertilization as well as by plant cultivar.
398
Ammonia-oxidising bacteria are considered to conduct the rate-limiting reaction of nitrification (Kowalchuk and
399
Stephen 2001). Soil and fertilization types exerted a different effect on nitrifying populations. Abundance of
N values of the same sampling date for the 0N treatment. Inorganic N fertilization significantly increased the abundance of
402
the amoA gene both in NA soil and in TO soil, thus confirming the findings by Chinnadurai et al. (2014). This finding is
403
consistent with the results of Kastl et al. (2015), who found that the abundance of amoA increased using inorganic fertilizer
404
and that the amoA abundance was positively correlated with increases in fertilization rates. The higher ammonia-oxidising
405
abundance in mineral N-fertilized than in compost-fertilized soils probably depended on the higher NH4+-N content of the
406
former soil (Chu et al. 2008). Higher amoA gene abundance and soil mineral N accumulated in the UREA than in the organic
407
manure-treated soil (Chu et al. 2007). The effect of inorganic N fertilization was significantly higher in TO soil than in NA
408
soil. The effect of inorganic N fertilization also demonstrated its propensity to produce mineral N, while the high rate of NA
409
soil nitrification was confirmed, even when a readily available source of N was applied to the soil.
410
Quantification of the abundance of denitrifiers may give insights on the denitrification activity and the potential N2O
411
fluxes (Henry et al. 2004). Real-time PCR assays, based on the amplification of nirK and nirS genes, provided a robust
412
estimation of changes in the abundance of denitrifiers (Yin et al. 2014). Denitrifier communities of both soils were dominated
413
by nirK, which suggests that in these agricultural soils, nitrification is mostly influenced by nitrite reductase encoded by this
414
gene. This fact may be used to predict NO emissions; indeed, the abundance of the nirK gene was positively correlated with
415
NO emission, while the abundance of the nirS gene was correlated with N2O emission (Szukics et al. 2009; Morales et al.
416
2010). Moreover, according to Yin et al. (2014), inorganic and organic fertilization have no influence on the abundance of
417
the nirS-denitrifier population. In contrast, the interaction between soil and fertilization types affected the abundance of
nirK-418
denitrifiers. Probably, NA soil had a higher content of anoxic micro-aggregates than the TO soil due to the clay texture. This
419
higher content of anoxic micro-aggregates may have favoured the activity of denitrifying bacteria. Organic C availability has
420
been shown to increase denitrification rates, especially in anoxic environments of aggregate structures that protect and
421
stabilize organic matter (Kong et al. 2010). In contrast, in TO soil, nirK gene abundance was probably related to N availability,
422
which increased with UREA inputs or when soil mineral N reached a plateau.
423
The initial delay phase of NA soil was probably due to the reduced tendency of this soil to produce nitric N and to
424
the increased tendency of this soil to denitrify due to the presence of anoxic sites.
425
Conclusions
426
427
Our experiments showed that there was a good correlation between chemical and microbiological measurements from
lab-428
soil incubation experiments and soil-plant N dynamics of maize cropping systems. Mineralization patterns proved to be useful
429
in optimizing the fertilization management of maize under field conditions as long as incubation time is normalized over the
maize growth cycle according to the simplified model of Growth Degree Days. According to this approach, the average
431
cumulative soil mineral N values, calculated over a short incubation period (42 days) covering the most sensitive phase of
432
crop N requirement, may be a good index of N availability to crops. Moreover, the shape and kinetic parameters of net N
433
mineralization from medium term (112 days) soil incubation provided consistent information on the interaction between
434
fertilizers and native fertility. The abundance of key genes of N cycle provides information about the soil attitude in making
435
N available for plant nutrition as affected by soil properties (i.e. abundance of ammonium fixing clays) and N availability
436
(mineral vs organic fertilization). Ourwork provides a prime starting point for the development of a crop-based approach for
437
using incubation data to manage the fertility of maize. Future research should aim to validate lab-derived N availability indices
438
with different maize cultivars (according to the FAO classification) and in different pedoclimatic conditions to extend our
439
approach to a wide range of maize-based cropping systems.
440
441
Aknowledgements
442
443
This publication is the result of MESCOSAGR project co-ordinated by Prof. A. Piccolo, and funded by the Italian Ministry
444
of Education, University and Research (MIUR), under the Special Integrative Funding for Research (FISR) initiative
445
(Strategic programme “Sustainable Development and Climatic Changes”). In alphabetical order, we thank Roberta
446
Buonomo, Adriana Impagliazzo, Roberto Maiello, Lucia Ottaiano and Rosa Russo for their technical assistance in
447
managing soil incubations.
448
References
450
451
Abbasi MK, Khizar A (2012) Microbial biomass carbon and nitrogen transformations in a loam soil amended with
organic-452
inorganic N sources and their effect on growth and N-uptake in maize. Ecol Eng 39:123–132
453
Allen RG, Pereira LS, Raes D, Smith M (1998) Crop evapotranspiration - Guidelines for computing crop water
454
requirements – FAO Irrigation and Drainage paper 56. FAO, Rome, Italy
455
Alluvione F, Fiorentino N, Bertora C, Zavattaro L, Fagnano M, Chiaranda FQ, Grignani C (2013) Short-term crop and soil
456
response to C-friendly strategies in two contrasting environments. Eur J Agron 45:114–123
457
Amlinger F, Götz B, Dreher P, Geszti J, Weissteiner C (2003) Nitrogen in biowaste and yard waste compost: dynamics of
458
mobilisation and availability–a review. Eur J Soil Biol 39:107–116
459
Annabi M, Houot S, Francou F, Poitrenaud M, Le Bissonnais Y (2007) Soil aggregate stability improvement with urban
460
composts of different maturities. Soil Sci Soc Am J 71:413–423
461
Braker GA, Fesefeldt A, Witzel KP (1998) Development of PCR primer systems for amplification of nitrite reductase
462
genes (nirK and nirS) to detect denitrifying bacteria in environmental samples. Appl Environ Microbiol 64:3769–3775
463
Chen YJ, Day SD, Wick AF, McGuire KJ (2014) Influence of urban land development and subsequent soil rehabilitation on
464
soil aggregates carbon and hydraulic conductivity. Sci Total Environ 494:329–336
465
Chen H, Mothapo NV, Shi W (2015) Soil moisture and pH control relative contributions of fungi and bacteria to N2O
466
production. Microbial Ecol 69:180–191
467
Chinnadurai C, Gopalaswamy G, Balachandar D (2014) Long term effects of nutrient management regimes on abundance
468
of bacterial genes and soil biochemical processes for fertility sustainability in a semi-arid tropical Alfisol. Geoderma
469
232:563–572
470
Chukwujindu MAI, Egun AC, Emuh FN, Isirimah NO (2006) Compost maturity evaluation and its significance to
471
agriculture. Pak J Biol Sci 9:2933–2944
472
Chu H, Fujii T, Morimoto S, Lin X, Yagi K, Hu J, Zhang J (2007) Community structure of ammonia-oxidizing bacteria
473
under long-term application of mineral fertilizer and organic manure in a sandy loam soil. Appl Environ Microb 73:485–
474
491
475
Chu H, Fujii T, Morimoto S, Lin X, Yagi K (2008) Population size and specific nitrification potential of soil
ammonia-476
oxidizing bacteria under long-term fertilizer management. Soil Biol Biochem 40:1960–1963
Cordovil CMdS, Coutinho J, Goss M, Cabral F (2005) Potentially mineralizable nitrogen from organic materials applied to
478
a sandy soil: Fitting the one-pool exponential model. Soil Use Manage 21:65–72
479
De Neve S, Hofman G (2002) Quantifying soil water effects on nitrogen mineralization from soil organic matter and from
480
fresh crop residues. Biol Fertil Soils 35:379–386
481
Delmont TO, Francioli D, Jacquesson S, Laoudi S, Nesme MJ, Ceccherini MT, Nannipieri P, Simonet P, Vogel TM (2014)
482
Microbial community development and unseen diversity recovery in inoculated sterile soil. Biol Fertil Soils 50:1069-1076
483
Diacono M, Montemurro F (2010) Long-term effects of organic amendments on soil fertility. A review. Agron Sustain Dev
484
30:401–422
485
Eriksen GN, Coale FJ, Bollero GA (1999) Soil nitrogen dynamics and maize production in municipal solid waste amended
486
soil. Agron J 91:1009–1016
487
Fagnano M, Adamo P, Zampella MV, Fiorentino N (2011) Environmental and agronomic impact of fertilization with
488
composted organic fraction from municipal solid waste. Agric Ecosyst Environ 141:100–107
489
Flavel TC, Murphy DV (2006) Carbon and nitrogen mineralization rates after application of organic amendments to soil. J
490
Environ Qual 35:183–193
491
Gabrielle B, Da-Silveira J, Houot S, Michelin J (2005) Field-scale modelling of carbon and nitrogen dynamics in soils
492
amended with urban waste composts. Agric Ecosyst Environ 110:289–299
493
Gaines TP, Gaines ST (1994) Soil texture effect on nitrate leaching in soil percolates. Commun Soil Sci Plant Anal
494
25:2561–2570
495
Griffin TS, He Z, Honeycutt CW (2005) Manure composition affects net transformation of nitrogen from dairy manures
496
Plant Soil 273:29–38
497
Hargreaves JC, Adl MS, Warman PR (2008) A review of the use of composted municipal solid waste in agriculture. Agric
498
Ecosyst Environ 123:1–14
499
Hayden HL, Drake J, Imhof M, Oxley APA, Norng S, Mele PM (2010) The abundance of nitrogen cycle genes amoA and
500
nifH depends on land-uses and soil types in South-Eastern Australia. Soil Biol Biochem 42:1774–1783
501
Henry S, Baudoin E, López-Gutiérrez JC, Martin-Laurent F, Brauman A, Philippot L (2004) Quantification of denitrifying
502
bacteria in soils by nirK gene targeted real-time PCR. J Microbial Meth 59:327–335
503
Holland JM (2004) The environmental consequences of adopting conservation tillage in Europe: reviewing the evidence.
504
Agric Ecosyst Environ 103:1–25
Huang CC, Chen ZS (2009) Carbon and nitrogen mineralization of sewage sludge compost in soils with a different initial
506
pH. Soil Sci Plant Nutr 55:715–724
507
Kandeler E, Deiglmayr K, Tscherko D, Bru D, Philippot L (2006) Abundance of narG nirS nirK and nosZ genes of
508
denitrifying bacteria during primary successions of a glacier foreland. Appl Environ Microb 72:5957–5962
509
Kastl E-M, Schloter-Hai B, Buegger F, Schloter M (2015) Impact of fertilization on the abundance of nitrifiers and
510
denitrifiers at the root–soil interface of plants with different uptake strategies for nitrogen. Biol Fertil Soils 51:57-64
511
Keeling AA, Cook JA, Wilcox A (1998) Effects of carbohydrate application on diazotroph populations and nitrogen
512
availability in grass swards established in garden waste compost. Bioresource Technol 66:89–97
513
Kondo M, Yasuda M (2003) Effects of temperature water regime light and soil properties on 15N
2 fixation associated with
514
decomposition of organic matter in paddy soils. Jpn Agric Res Q 37:113–119
515
Kong AYY, Hristova K, Scow KM, Six J (2010) Impacts of different N management regimes on nitrifier and denitrifier
516
communities and N cycling in soil microenvironments. Soil Biol Biochem 42:1523–1533
517
Kowalchuk GA, Stphen JR (2001) Ammonia-oxidizing bacteria: a model for molecular microbial ecology. Ann Rev
518
Microbiol 55:485–529
519
Lindsay EA, Colloff MJ, Gibb NL, Wakelin SA (2010) The abundance of microbial functional genes in grassy woodlands is
520
influenced more by soil nutrient enrichment than by recent weed invasion or livestock exclusion. Appl Environ Microb
521
76:5547–5555
522
Lynch DH, Voroney RP, Warman PR (2006) Use of 13 C and 15 N natural abundance techniques to characterize carbon
523
and nitrogen dynamics in composting and in compost-amended soils. Soil Biol Biochem 38:103–114
524
Meng Q, Sun Y, Zhao J, Lianren Z, Xianfa M, Meng Z, Wei G, Guangcheng W (2014) Distribution of carbon and nitrogen
525
in water-stable aggregates and soil stability under long-term manure application in solonetzic soils of the Songnen plain
526
northeast China. J Soils Sediments 14:1041–1049
527
Miller JJ, Beasley BW, Drury CF, Zebarth BJ (2010) Available nitrogen and phosphorus in soil amended with fresh or
528
composted cattle manure containing straw or wood- chip bedding. Can J Soil Sci 90:341–354
529
Morales SE, Cosart T, Holben WE (2010) Bacterial gene abundances as indicators of greenhouse gas emission in soils.
530
ISME J 4:799–808
531
Nedvěd V, Balík J, Černý J, Kulhánek M, Balíková M (2008) The changes of soil nitrogen and carbon contents in a
long-532
term field experiment under different systems of nitrogen fertilization. Plant Soil Environ 54:463–470
Nyamangara J, Piha MI, Kirchmann H (1999) Interactions of aerobically decomposed cattle manure and nitrogen fertilizer
534
applied to soil. Nutr Cycl Agroecosyst 54:183–188
535
Orr CH, James A, Leifert C, Cooper JM, Cumming SP (2011) Diversity and activity of free-living nitrogen-fixing bacteria
536
and total bacteria in organic and conventionally managed soils. Appl Environ Microb 77:911–919
537
Pampana S, Ercoli L, Masoni A, Arduini I (2009) Remobilizaiton of dry matter and nitrogen in Maize as affected by hybrid
538
maturity class. Ital J Agron 2:39–46
539
Pepe O, Ventorino V, Blaiotta G (2013) Dynamic of functional microbial groups during mesophilic composting of
agro-540
industrial wastes and free-living (N2)-fixing bacteria application. Waste Manage 33:1616–1625
541
Pereira E, Silva MC, Schloter-Hai B, Schloter M, van Elsas JD, Salles JF (2013) Temporal dynamics of abundance and
542
composition of nitrogen-fixing communities across agricultural soils. PloS One 8 e74500. doi:
543
10.1371/journal.pone.0074500
544
Piccolo A, Spaccini R, Nieder R, Richter J (2004) Sequestration of a biologically labile organic carbon in soils by humified
545
organic matter. Clim Chang 67:329–343
546
Poly F, Monrozier LJ, Bally R (2001) Improvement in the RFLP procedure for studying the diversity of nifH genes in
547
communities of nitrogen fixers in soil. Res Microbiol 152:95–103
548
Ribeiro HM, Fangueiro D, Alves F, Ventura R, Coelho D, Vasconcelos E, Cunha-Queda C, Coutinho J, Cabral F (2010)
549
Nitrogen mineralization from an organically managed soil and nitrogen accumulation in lettuce. J Plant Nutr Soil Sci
550
173:260–267
551
Richards LA (1941) A pressure-membrane extraction apparatus for soil solution. Soil Sci 51:377–386
552
Rodríguez-Blanco A, Sicardi M, Frioni L (2015) Plant genotype and nitrogen fertilization effects on abundance and
553
diversity of diazotrophic bacteria associated with maize (Zea mays L.). Biol Fertil Soils 51:391-402
554
Rösch C, Mergel A, Bothe H (2002) Biodiversity of denitrifying and dinitrogen-fixing bacteria in acid forest soil. Appl
555
Environ Microbiol 68:3818–3829
556
Rotthauwe JH, Witzel KP, Liesack W (1997) The ammonia monooxygenase structural gene amoA as a functional marker:
557
molecular fine-scale analysis of natural ammonia-oxidizing populations. Appl Environ Microbiol 63:4704–4712
558
Roper M, Smith N (1991) Straw decomposition and nitrogenase activity (C2H2 reduction) by free-living microorganisms
559
from soil: Effects of pH and clay content. Soil Biol Biochem 23:275–283
560
Saenger A, Geisseler D, Ludwig B (2014) C and N dynamics of a range of biogas slurries as a function of application rate
